Search

Understanding local-scale paleoecological processes and their driving mechanisms is a major challenge in plant community ecology. However, phytolith records have the potential to retrieve localized signals of the evolution and patterns of past ecosystems. As a case study, we selected a forested region in Northeast China. We first established a modern phytolith reference dataset which enabled numerical estimates to be made of various aspects of the plant community. We then apply the dataset to a phytolith record from a sedimentary profile (HN) from the Changbai mountain region in order to quantitatively reconstruct the evolution of the vegetation since the last deglaciation. The results of partial least squares regression were used demonstrate that modern phytoliths can quantitatively estimate plant community coverage, species diversity and dry weight ratio; three numerical estimates respectively explained 62.3%, 50.3% and 49.5% of the variance. These quantitative relationships can provide reliable modern phytolith analogues for paleoecological reconstructions. The modern phytolith-based vegetation reconstruction indicates that the Changbai mountain region was dominated by forest communities since the late glacial period. The temporal development of the vegetation can be summarized as follows: boreal or coniferous forest, sparse and species-poor → mixed broadleaf-conifer forest, relatively dense and species-rich → mixed broadleaf-conifer forest, dense and species-rich. The phytolith-inferred changes in plant community coverage, dry weight ratio and species diversity correspond well with climatic changes. This supports the view that the environmental background mainly dominates ecosystem evolution, and it also confirms the utility of phytolith records to resolve local paleovegetation patterns and evolution. [ABSTRACT FROM AUTHOR]